Hydraulic system reliability modelling and analysis based on convex model Bayesian network

Compared with the probabilistic method, the non-probabilistic convex model only requires a small number of samples and experiences to obtain the variation bounds of the ambiguous and imprecise parameters, and whereby makes the reliability analysis very convenient. In this paper, hydraulic lifting system reliability model is established and the reliability of the system is analyzed by convex model Bayesian network, then the reliability indices are described in the form of interval parameters. Using ellipsoid model to constrain interval model, the more accurate reliability indices are obtained. Finally the convex model Bayesian network is compared with traditional Bayesian network. All the reliability indices obtained by convex model Bayesian network provide the basis for condition-based maintenance, fault diagnosis and reliability design.